Agentic coding risks increasing technical debt and system complexity
Svix CEO Tom Peck-Owen warns that rapid AI code generation leads to unsustainable complexity, akin to offshoring. He argues that intentional simplicity and deterministic checks are required to manage the resulting technical debt.
Svix CEO Tom Peck-Owen and host Conor Bronsden discussed the impact of agentic coding on software engineering practices. They noted that while AI agents can generate code rapidly, this often leads to 'patches on top of patches' and increased complexity, similar to the offshoring model of adding more engineers to a problem. Peck-Owen cited a friend's argument that this approach may eventually reverse due to technical debt, though he believes the scale of AI generation is infinitely larger than previous offshoring. The discussion highlighted that current agentic workflows often prioritize speed over simplicity, resulting in codebases that are harder to maintain. Tools like T3 Code are introducing architectural animation to visualize data flow and identify vestigial code, while developers use deterministic workflows and agents to monitor systems like Datadog to catch errors. The consensus is that simplicity is becoming a critical skill to prevent agentic flows from exacerbating existing infrastructure problems.